【Python】用bs4解析网页3

NoviceLearningRecord

已于 2023-02-16 09:17:05 修改

阅读量275

点赞数

分类专栏： Python 文章标签： python 开发语言

于 2023-02-15 13:22:59 首次发布

本文链接：https://blog.csdn.net/weixin_47542662/article/details/129041563

版权

Python 专栏收录该内容

16 篇文章 0 订阅

订阅专栏

1、注意list是全局变量还是局部变量，不然存储的内容可能是重复的；

2、命名规则要规范一下，写得时候就按照顺序来，不然后续同名的太多，很容易报错，找bug很麻烦；

3、如果有爬出的内容有编码错误，要encoding一下；

#     headers = {'User-Agent':'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/106.0.0.0 Safari/537.36'}
#     response = requests.get(url = path,headers = headers)
#     # 设置编码格式，不设置就是乱码
#     response.encoding = "utf-8"
#     response = response.text

4、用driver.get()很有用，基本上都能爬取出来，就是比较慢一些；

    driver = webdriver.Chrome()
    # 向指定url发起请求
    driver.get(path)
    # 刷新页面,防止页面未渲染
    driver.refresh()
    # 等待页面渲染
    time.sleep(2)
    # 获取网页源代码
    html = driver.page_source
    # 退出模拟浏览器，防止进程残留
    driver.quit()
    soup = BeautifulSoup(html,'html.parser')

这段代码写了几个循环就是为了一层层获得要爬取网页的链接，这个链接直接获取不到，只能一层层拿到；

# 中国银行间市场交易商协会
import csv
import time
import requests
from bs4 import BeautifulSoup
import os
def get_html(path):
    driver = webdriver.Chrome()
    # 向指定url发起请求
    driver.get(path)
    # 刷新页面,防止页面未渲染
    driver.refresh()
    # 等待页面渲染
    time.sleep(2)
    # 获取网页源代码
    html = driver.page_source
    # 退出模拟浏览器，防止进程残留
    driver.quit()
    soup = BeautifulSoup(html,'html.parser')
    return soup

if __name__=='__main__':
    start = time.time()
#     第一层链接
    path_1 = r'https://www.nafmii.org.cn/yj/jrscyj/xswyh/202112/t20211214_116082.html'
    soup = get_html(path_1)
    link_info_1 = soup.find_all('a')[30:39]
#     print(link_info_1)
#     link_list_1用来存第二层链接
    link_list_1 = []
    for link in link_info_1:
        link_1 = link.get('href')
        print(link_1)
#         title  =link.text
#     存在字典里可能更方便一些
#     link_list[title] = link1
#     print(title)
        link_list_1.append(link_1)
    print('link_list_1:%s'%link_list_1)
#      根据第二层链接获得第三层链接
    link_list_2 = []
    for link2 in link_list_1:
        soup1  = get_html(link2)
        link_info_2 = soup1.find_all('div',class_='text-cont')
# 得到链接，先缩小范围，然后进一步解析
        for row_1 in link_info_2:
            link_2 = row_1.find_all('a')[0]
            link_2 = link_2.get('href').replace('./','')
            #         print(type(link_1))
            path2 = os.path.join(link2,link_2)
            link_list_2.append(path2)   
    print('link_list_2:%s'%link_list_2)
#     用来盛放第三层链接,也就是我们要爬取网页的链接
    link_list_3 = []
    for link3 in link_list_2:
        soup2  = get_html(link3)
        link_info_3 = soup2.find_all('div',class_='yh_sub_textList ty_list')
    #     print(link_info2)
    #     <div class="yh_sub_textList ty_list">
        for row_2 in link_info_3:
            row_2_1 = row_2.find_all('a')[0]
            link4 = row_2_1.get('href').replace('./','')
    #         print(link4)
            final_list = os.path.join(link3,link4)
            link_list_3.append(final_list)
    print(link_list_3)
    fp = open(r'D:\test\zhognguoyinhangjianshichang1.csv','w',encoding = 'utf-8',newline='')
    for final_link in link_list_3:
        final_info = []
        soup3  = get_html(final_link)
        title = soup3.find_all('h1')[0].text+'\n'
        fp.write(title)
        main_info = soup3.find_all('div',class_="yh_viewTextBox")
    #     <div class="yh_viewTextBox">
        for row_3 in main_info:
            row_3_1 = row_3.find_all('tr')
            for row_3_2 in row_3_1:
                row_3_3 = row_3_2.find_all('td')
#                 因为有第二列和第三列为空的情况，所以要分开写；
                j_title = row_3_3[0].text.replace('\n','')
                if len(row_3_3)==1:
                    name = ''
                    final_info.append([j_title,name])
                if len(row_3_3)==2:
                    name = row_3_3[1].text.replace('\n','')
                    final_info.append([j_title,name])
                if len(row_3_3)==3:
                    name = row_3_3[1].text.replace('\n','')
                    col3 = row_3_3[2].text.replace('\n','')               
                    final_info.append([j_title,name,col3])
    # print(final_info)
        writer = csv.writer(fp)
        writer.writerows(final_info)
    fp.close()
    end = time.time()
    r_time = end-start
    print(r_time)
    print('finished!')

发现我挨个写td的做法不太简洁，完全没必要，直接用for循环就行，本质上也是用列表存列表，真傻我

# 中国上市协会行业委员会
import csv
import time
import os
from bs4 import BeautifulSoup 
# 提取html
def get_html(path):
    driver = webdriver.Chrome()
    driver.get(path)
    time.sleep(3)
    driver.refresh()
    html  = driver.page_source
    driver.quit()
    soup = BeautifulSoup(html,'html.parser')
    return soup
if __name__=='__main__':
#     第一层链接
#     path = 'https://www.capco.org.cn/xhdt/hyzywyh/cwzjzywyh/index.html'
#     soup  = get_html(path)
# #     print(soup)
#     soup  = soup.find_all('ul',class_='navList')
# #     新建一个list，用来放第一层链接
#     url_list = []
# #    把第一个委员会的链接接存进去，方便一起操作
#     url_list.append(path)
# #     第一个链接不用放，因为这个就是原来的链接，所以从第二个链接开始存储
#     for row in soup:
#         row = row.find_all('a')[1::]
#         for url in row:
#             link  = url.get('href').replace('../','')
#             pinjie = 'https://www.capco.org.cn/xhdt/hyzywyh/'
#             fir_link = os.path.join(pinjie,link)
#             url_list.append(fir_link)
# #             print(fir_link)
#     print(url_list)
# #     爬取第二层链接，这里面有个问题需要注意一下，如何判断委员会名单的信息
#     for link1 in url_list:
#         soup1 = get_html(path)
#         link_info = soup1.find_all('ul',class_='listWrap mt20')
#         for row in link_info:
#             print(row.text)
    s = time.time()
    list1 = ['https://www.capco.org.cn/hyxx/fhz/index.html','https://www.capco.org.cn/hyxx/jsh/jshcy/','https://www.capco.org.cn/xhdt/hyzywyh/cwzjzywyh/201909/20190920/j_2019092017422400015689725955897856.html','https://www.capco.org.cn/xhdt/hyzywyh/dldswyh/201904/20190418/j_2019041817540900015689727295348545.html','https://www.capco.org.cn/xhdt/hyzywyh/bgrzwyh/201909/20190920/j_2019092017483500015689729664587942.html','https://www.capco.org.cn/xhdt/hyzywyh/dshmswyh/201909/20190920/j_2019092017501500015689730664697968.html','https://www.capco.org.cn/xhdt/hyzywyh/jshzywyh/201909/20190920/j_2019092017513300015689731443997993.html','https://www.capco.org.cn/xhdt/hyzywyh/tzzglwyh/201909/20190920/j_2019092017524700015689732182128018.html','https://www.capco.org.cn/xhdt/hyzywyh/xsgwwyh/202206/20220607/j_2022060714184200016545827360056820.html','https://www.capco.org.cn/xhdt/hyzywyh/szw/202206/20220622/j_2022062221114900016559035224143357.html','https://www.capco.org.cn/xhdt/hyzywyh/xnyyznqcwyh/202210/20221028/j_2022102809352000016669209337028831.html']
    print(list1)
    final_info = []
    for url in list1:
        soup  = get_html(url)
        soup = soup.find_all('div',class_='artical')
        for row in soup:
            title = row.find_all('h3')[0].text+'\n'
            final_info.append([title])
            print(title)
            main_info = row.find_all('tr')       
            for cell in main_info:
                cell = cell.find_all('td')
                main = []
                for td in cell:         
                    td = td.text
#                     print(td)
                    main.append(td)
                final_info.append(main)
    fp = open(r'D:\test\shangshixiehui-xiehui.csv','w',encoding = 'utf-8',newline='')
    writer = csv.writer(fp)
    writer.writerows(final_info)
    e = time.time()
    run = e-s
    fp.close()
    print('runtime is %d'%run)
    print('finished!')